Strength for one week, No Charts Attached ;) | Page 9 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Charts, Charts, Charts! 1 reply

Hallett Norman - Mental Fitness For Traders (Attached) 2 replies

EA works attached to single chart, messes up with multiple charts 2 replies

The Pocket Book Of Economic Indicators (Attached) 2 replies

One More Zero - How To Trade FX Like A Pro (Attached) 0 replies

Attachments: Strength for one week, No Charts Attached ;)
Exit Attachments

Strength for one week, No Charts Attached ;)

  • #161
  • Edited 3:20am Aug 11, 2011 3:03am | Edited 3:20am
Yesterday started to code a little program, which can help test the system semi-manually. Maybe today evening will sit again to it.

Entering one week interval, opens orders, it goes through five days, everyday (on setted time), calculates negative positions, close half and shows pips sum at the end of the week.

Same process with another week.
Can do it more automatically, but can't find broker offering all 28 pairs.


Donatas
  • #162
  • Aug 11, 2011 3:44am Aug 11, 2011 3:44am
well my basket from last night is not doing well, nearly 3000 pips in the hole

i think this will be the major issue, dealing with correction days
  • #163
  • Aug 11, 2011 3:53am Aug 11, 2011 3:53am
I too started a basket last night based on last 10 days. I was using 1.0 lot size on a £50k demo account. So far I'm about 9k down having closed the five biggest losers.

I looks like the biggest losing pairs (and I only have 4 in profit) all had major turns at around midnight.

Glad it wasn't a live account.

GG
Now trading live
  • #164
  • Aug 11, 2011 4:06am Aug 11, 2011 4:06am
Quoting BaliuxXx
Yesterday started to code a little program, which can help test the system semi-manually. Maybe today evening will sit again to it.

Entering one week interval, opens orders, it goes through five days, everyday (on setted time), calculates negative positions, close half and shows pips sum at the end of the week.

Same process with another week.
Can do it more automatically, but can't find broker offering all 28 pairs.


Donatas
This would be really usefull, thank you
  • #165
  • Aug 11, 2011 4:36am Aug 11, 2011 4:36am
Quoting Gg33
I too started a basket last night based on last 10 days. I was using 1.0 lot size on a £50k demo account. So far I'm about 9k down having closed the five biggest losers.

I looks like the biggest losing pairs (and I only have 4 in profit) all had major turns at around midnight.

Glad it wasn't a live account.

GG
It is quite logical that if trading all currencies towards their previous strength then there will be a big loss as soon as we have retracement or reversal. Question is how many winning days and how many losing days we have during a longer period. One day or one week is not important at all.

BTW: My today's basket is also in heavy minus but it is still twice smaller than yesterday's profit. Let us see if the profit will be swiped off totally
  • #166
  • Aug 11, 2011 4:48am Aug 11, 2011 4:48am
I think that money management is very important and it can`t be traded with same lot size. I have an indicator that calculates the lot size per trade according to ATR. So due to different volatility and diferent pip value it gives diferent lot size. Just take a look at the charts usd/cad would trade 0.05 size while gbp/chf only 0.01 and the diference is huge not to take atention. I set 1 persent risk per trade and ATR period 30 on weekly chart.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 1.gif Size: 74 KB
Click to Enlarge Name: 2.gif Size: 77 KB
  • #167
  • Aug 11, 2011 4:52am Aug 11, 2011 4:52am
here is the indicator for calculation of risk management.
Attached File(s)
File Type: mq4 SaintSurferRiskMM.mq4   2 KB | 446 downloads
  • #168
  • Aug 11, 2011 4:59am Aug 11, 2011 4:59am
I think this strategy might be overall profitable in the long term. However, the question becomes... is it profitable enough to make it worth the while? It might not produce enough profits in the end to outperform other potential strategies. I envision periods of time where price will range amongst most of the traded pairs. Now the losses might not be much during a ranging week, but the time is wasted. Then you have periods where the currencies reverse trend. These periods will have large losses. Or worst of all, a 2-3 week correction. These periods will be really bad because you will lose during the correction and then lose again when trend resumes because strength measure falsely indicates a trend reversal. Remember, we are taking historical information to try and predict future direction. It's not much different than trading in the direction of trends, except you are comparing the trends of all the currencies against one another.

Again it might be profitable over the stretch of a year (or years), but probably nothing out of this world and maybe not even better than other strategies you could have been trading instead. Regardless, a thank you is in order for the creator of this thread for presenting the strategy and sharing it. Thank you FB.
  • #169
  • Aug 11, 2011 5:01am Aug 11, 2011 5:01am
of course, retracements you would expect to have heavy losses, these should be offset by the wins

the only problem i see is if you start trading it on a live acct. and are unlucky enough to start during one of these retracement periods, ie. you get all the losses before any gains..

with good mm this shouldn't be to much of a problem, unless of course the retracement turns into a trend change, the 10week period would have such a lag that it could be possible to have 2 to 3 weeks of heavy losses

by the way, happy to report that my trades are at least moving back in the correct direction now only 2000 pips down

Quoting Indrek
It is quite logical that if trading all currencies towards their previous strength then there will be a big loss as soon as we have retracement or reversal. Question is how many winning days and how many losing days we have during a longer period. One day or one week is not important at all.

BTW: My today's basket is also in heavy minus but it is still twice smaller than yesterday's profit. Let us see if the profit will be swiped off totally
  • #170
  • Aug 11, 2011 6:49am Aug 11, 2011 6:49am
hello all, and thanks for sharing this idea.

please, what broker are you using that allows all this pairs in demo? whit my broker only allow me in demo 12-13 pairs (alpari UK)

in real there are 26 crosses of this basket, but i am not going to do that for the moment

and also, are you noticed the big diference whit the spread between demo and real???

this is in effect a great problem for example eun/nzd 15 pips gbp/nzd 10 pips aud/nzd 12 pips and most of the crosses has 7-8 pips spred only 6-7 pairs has reasonable spread under3-4 pips, this supposes that in real mode we have a net loss aprox of 200 pips when we open all the positions.

thank you guys and plase tell me what broker are you using.
  • #171
  • Aug 11, 2011 6:53am Aug 11, 2011 6:53am
I'd like to share following link with you.
It's a tool to calculate margin and pip value.

In order to balance the sytem more we should adapt lotsizes according to pip value. Pip value should be the same across all pairs at the start of a new week.

Pip / Margin calculator

I also noticed some of you are asking: "What's the point in opening multiple CHF pairs in the same direction?"

Well, if you think of trading the different pairs as betting on "horses" then strength shows us what horse is likely to beet another horse...

E.a. Horse JPY has been winning from horses USD and EUR, but has been loosing against horse CHF... Get the picture?

Well if you where betting horses you would probably be betting more money on the strongest horse as compared to weaker horses...

As a currencie is around the middle of strength it gets harder to predict if it will win or loose the next race (trading week ). Therefore you would want to spread your bets on this currencie... in such a way that it might win against certain currencies but also might loose against other currencies.

Hope this comes across ... I think it's part of the system to trade all combinations of the 8 major pairs.

If you want to control risk... And offcourse we want that! It makes more sence to adjust the overall lotsize then to eliminate pairs... If you do eliminate pairs you might miss profit that was needed to offset losses on other pairs...

In order to calculate appropriate overall lotsize it might be worthwhile to calculate total volatility across all currencies and use this to calculate lotsizes for a certain amount of equity... so.. anybody an idea on how to do this?

Owh... and about trading daily... It seems to me that trading this system based on strength of last 10 days each day makes it much more vulnerable to market "noise", news events, large players, as opposed to looking at the weekly timeframe...

I think on a weekly bases, results will be much more stable...

See you later!
  • #172
  • Aug 11, 2011 7:30am Aug 11, 2011 7:30am
Alpari UK has all the crosses, but by default they are not all shown

To show all the crosses get up the Market Watch window, right click, then select Show All



Quoting anama
hello all, and thanks for sharing this idea.

please, what broker are you using that allows all this pairs in demo? whit my broker only allow me in demo 12-13 pairs (alpari UK)

in real there are 26 crosses of this basket, but i am not going to do that for the moment

and also, are you noticed the big diference whit the spread between demo and real???

this is in effect a great problem for example eun/nzd 15 pips gbp/nzd 10 pips aud/nzd 12 pips and most of the crosses has 7-8 pips spred only 6-7 pairs has reasonable spread under3-4 pips, this...
  • #173
  • Aug 11, 2011 7:45am Aug 11, 2011 7:45am
Quoting FBForex
In order to calculate appropriate overall lotsize it might be worthwhile to calculate total volatility across all currencies and use this to calculate lotsizes for a certain amount of equity... so.. anybody an idea on how to do this?
I suggest math of elementary school level
We have to multiply pip value of all pairs to their volatility, and then we easily find appropriate lot sizes by comparing the values.

Sample:
Pair 1 - pip value 1, ATR 150 -> 1 * 150 = 150
Pair 2 - pip value 1.3, ATR 100 - > 1.3 * 100 = 130
So, if you trade Pair 1 with 0.1 lots then you trade Pair 2 with 150/130*0.1=0.12 lots
  • #174
  • Aug 11, 2011 8:24am Aug 11, 2011 8:24am
thank you veru much, jmw. now i have it.
  • #175
  • Aug 11, 2011 8:39am Aug 11, 2011 8:39am
Quoting Indrek
I suggest math of elementary school level
We have to multiply pip value of all pairs to their volatility, and then we easily find appropriate lot sizes by comparing the values.

Sample:
Pair 1 - pip value 1, ATR 150 -> 1 * 150 = 150
Pair 2 - pip value 1.3, ATR 100 - > 1.3 * 100 = 130
So, if you trade Pair 1 with 0.1 lots then you trade Pair 2 with 150/130*0.1=0.12 lots
I see your point, however wouldn't it be more logical to use overall ATR (average ATR over all pairs) as an indication of total market volatility?

If we use pair specific volatility we are increasing risk on slow moving pairs... which probably have less directional movement then faster moving pairs (probably pairs with very strong vs very weak currencie).

We want to trade higher risk on pairs with more chance on movement in trend direction.

So using overall ATR on all pairs might be the solution?
  • #176
  • Aug 11, 2011 8:54am Aug 11, 2011 8:54am
please i want everyone that are already demoing this system to post their results in compare to the past (especially yesterdays result). i can see a lot of move on chf pairs today contrary to the previous trend direction as someone forcasted before. Like EChf moved over 500 pips. Just want to know if the correlated pairs still able to adjust to the drawdown. Thanks
We Win Some, We Loose Some, But Let your Win Rates be More to be Successful
  • #177
  • Aug 11, 2011 8:59am Aug 11, 2011 8:59am
Quoting adeforex
please i want everyone that are already demoing this system to post their results in compare to the past (especially yesterdays result). i can see a lot of move on chf pairs today contrary to the previous trend direction as someone forcasted before. Like EChf moved over 500 pips. Just want to know if the correlated pairs still able to adjust to the drawdown. Thanks
In such days it is quite obvious that the system fails. Because we do not use correlations like we do when hedging. We trade all currencies towards there strength, so we are either totally right or totally wrong about majority of pairs in basket
Question is if it will kill all earlier profit and what is the result for a longer period like 6 months.
  • #178
  • Aug 11, 2011 9:04am Aug 11, 2011 9:04am
using a daily method does not seem like a good idea, like FBForex says the weekly might be better suited.

ok, so I was probably using too larger lotsize (0.1 on a 5kUSD acct) but my basket from yesterday blew out the acct.

will try next week on monday again but based on the 10 weekly as originally suggested
  • #179
  • Aug 11, 2011 9:14am Aug 11, 2011 9:14am
Quoting jmw1970
using a daily method does not seem like a good idea, like FBForex says the weekly might be better suited.

ok, so I was probably using too larger lotsize (0.1 on a 5kUSD acct) but my basket from yesterday blew out the acct.

will try next week on monday again but based on the 10 weekly as originally suggested
If there is big intervention or sth (like today) then your account will be gone on weekly trades too if your lot sizes are too big.
Of course, when taking weekly trades you can make five times less mistakes And you do not have to buy CHF all the time at the time when it is falling.
Let us see what guys will have for this week's result. CHF is very very bad now...
  • #180
  • Aug 11, 2011 10:41am Aug 11, 2011 10:41am
Quoting zznbrm
I've backtested for the past 10 weeks. Tried to follow Post 1 rules, but I did use my own calculations for the percent change. I did this since I do not know the exact calculation the Forex Army website uses. They could be throwing darts at a dartboard for all we know.

It really goes without saying...results will vary greatly depending on the percent change calculations, how far back you go for your reference prices, and when/how often you cut losing trades.
Here is a summary of 5-month automated backtest. Results are in US dollars, trading on a FxPro standard account using 0.1 lots (10,000 units) per trade. FxPro is on Eastern European time, so I adjusted the times accordingly. Account started with $10,000 and, as you can see, ended with $13,655.63. I used very conservative spreads, but did NOT include swap.

During the week of 5/9/2011, drawdown got to about 33%.

If anyone is interested in the full results, I will post the entire result set. Like I said previously, my only deviation from Post 1 rules was the currency percentage change calculation.
Attached Image
0 traders viewing now